Drug addiction is a debilitating disease that causes considerable social, psychological, and financial harm. The hallmark symptom of addiction, an intense desire to obtain and consume drugs, is often triggered by drug-associated sensory cues. The mechanism that gives rise to these symptoms relies on drug-induced increase in dopamine--a neurochemical normally involved in associative learning, reward-processing, and motivation--that disproportionately reinforces drug-paired sensory experience. In order to design therapies for preventing and treating substance abuse, it is crucial to determine the neural mechanisms that govern natural dopamine-mediated reinforcement and its perturbation in addiction. The brain region that is most critically affected by drugs of abuse is the ventral striatum, and it is within the ventral striatum that dopamine attributes neutral sensory experience with rewarding, motivational properties. This dopamine-dependent transformation involves changes in neural activity in large populations of neurons that belong to several functionally distinct classes. However determining the exact nature of these changes--and how they underlie the emergence of cue-triggered behaviors in addiction--has not been possible due to the inaccessibility of the ventral striatum to large-scale, high-resolution monitoring of neural activiy. To address this limitation and identify the dopamine- mediated transformation that renders drug-paired sensory cues highly motivating, we propose to deploy a unique surgical approach we have developed in the anesthetized mouse that will for the first time enable large- scale monitoring of activity in functionally identified neurons in the ventral striatum. Using this approach, coupled with multiphoton population imaging we will, in AIM I, determine general principles of sensory encoding in a subdivision of the ventral striatum, and the specific ways in which this encoding is modified by acute administration of cocaine and amphetamine. Then in AIM II, we will address how natural rewards and drugs of abuse alter neural representations of sensory experience by coupling our imaging approach with traditional classical conditioning approaches. The proposed project will be the first to observe general principles of sensory encoding in the ventral striatum in vivo by monitoring activity in large populations of genetically identified neurons belonging to functionally distinct classes. Together, the proposed experiments will reveal what features of this activity change as neutral sensory experience acquires motivational value through natural-reward learning and how this transformation underlies striatal principles of associative learning that are perturbed by addiction.